Previous research has demonstrated that while conditioned fear may rely predominantly on the central nucleus of the amygdala (CEA), responses to less predictable or unconditioned anxiogenic stimuli, or fear states akin to anxiety may depend on the bed nucleus of the stria terminalis (BNST). Furthermore, this dissociation may be related to CRF type I (CRF-RI) receptor activity in the two regions. The major goal, and second specific aim of this proposal is to use a lentiviral vector to over-express the CRF-R1 receptor in the BNST or CEA of adult rats and assess its effects on different forms of facilitation of the acoustic startle reflex, a measure of fear. The hypothesis is that over-expression in the CEA will facilitate fear-potentiated startle; a form of conditioned fear, and over-expression in the BNST will facilitate baseline acoustic and light-enhanced startle, measures of general stress/anxiety and unconditioned fear, respectively. These studies assume that over-expression of a receptor is rate limiting. They also depend on a localized and sustained expression of the CRF-R1 transgene. Therefore, the first specific aim is to a) assess whether viral mediated increases in CRF-R1 receptors will change binding and cAMP levels in vitro, B) characterize in-vivo the spread of virus following injection and C) characterize the time course of expression. The proposed studies aim to contribute to our understanding of the neurological basis of anxiety disorders or disorders of fear (e.g. post-traumatic stress disorder).